The is-ought problem was articulated by the Scottish philosopher David Hume in Book III, Part I, Section I of A Treatise of Human Nature (1739–40). In a now-canonical passage, Hume observes that writers on morality proceed for some time with ordinary reasoning about what is and is not the case, and then, "of a sudden," substitute propositions connected by ought or ought not for those connected by is or is not. Hume insists that this transition is "imperceptible" yet of the "last consequence," because the new relation expressed by ought would need to be explained and justified — and he cannot see how it could be deduced from premises "entirely different from it." The challenge does not appear under the label "is-ought problem" in Hume's own text; that name, along with the vivid metaphor Hume's Guillotine (the clean logical cut severing facts from values), was popularised by later commentators, notably the German philosopher Max Black in his 1964 essay "The Gap Between 'Is' and 'Should'."
Mechanically, the problem is a constraint on deductive inference. In a valid deductive argument, the conclusion can contain no term or relation that does not appear, at least implicitly, in the premises. If every premise is a descriptive statement of fact — that an action causes suffering, that a contract was signed, that a population is starving — then a conclusion containing the evaluative operator "ought" introduces a new element with no warrant in the premises. The argument is therefore formally invalid. To restore validity, the reasoner must insert at least one premise that is itself normative or evaluative — for example, "one ought not to cause unnecessary suffering." Hume's point is that this bridging premise must come from somewhere other than observation of the world; for Hume himself, it derives from sentiment, the "passions" and moral approbation rooted in human feeling rather than reason.
The problem admits several refinements and adjacent formulations that practitioners should keep distinct. G. E. Moore's naturalistic fallacy, advanced in Principia Ethica (1903), is a related but separate thesis: Moore argued that "good" is a simple, non-natural property that cannot be defined in terms of any natural property such as pleasure or desire. The is-ought gap is a claim about the logic of inference; the naturalistic fallacy is a claim about the analysis of moral concepts. A further variant is the "fact-value distinction," which the logical positivists radicalised into emotivism — the view, associated with A. J. Ayer's Language, Truth and Logic (1936) and C. L. Stevenson, that moral statements are not truth-apt at all but expressions of attitude. These are three distinct theses that are frequently, and incorrectly, conflated.
The structure recurs constantly in contemporary policy and diplomatic argument. When the Intergovernmental Panel on Climate Change publishes its assessment reports — for example the Sixth Assessment Report finalised in 2023 — its working groups are careful to separate descriptive findings about warming trajectories from the prescriptive question of what mitigation targets states ought to adopt, a boundary the IPCC itself frames as the line between science and "policy-relevant but not policy-prescriptive" advice. Similarly, when a foreign ministry's intelligence assessment states that a regime possesses a capability, that finding alone cannot generate the conclusion that intervention is warranted; an additional normative premise about just cause must be supplied. The 2003 Iraq debate turned in part on conflating an "is" (alleged weapons) with an "ought" (the duty to disarm by force).
The is-ought problem must be distinguished from the closely adjacent concept of moral realism and from the genetic fallacy. Moral realism — the position that there exist objective moral facts — is not refuted by Hume's Guillotine; a realist can accept that moral facts are not derived from non-moral facts while still holding that they exist sui generis. The is-ought gap is also narrower than the broader claim that reason plays no role in ethics: Hume permits reason to inform us about consequences and means, leaving the determination of ends to sentiment. Confusing the inferential gap with wholesale moral scepticism is a common analytical error.
Several philosophers have contested the guillotine's sharpness. John Searle, in "How to Derive 'Ought' from 'Is'" (1964), argued that institutional facts about promising can bridge the gap: uttering "I promise" places one under an obligation, allegedly yielding an "ought" from an "is." Critics, including R. M. Hare, responded that Searle smuggles in a constitutive normative rule. Philippa Foot and the later Aristotelian naturalists (Foot's Natural Goodness, 2001) argued that facts about a creature's flourishing already carry evaluative force, dissolving the dichotomy. The debate remains unresolved, and the guillotine endures as a methodological caution rather than a proven theorem.
For the working practitioner — the civil servant drafting policy, the diplomat justifying a position, the UPSC aspirant preparing the GS Paper IV ethics syllabus — the is-ought problem is an indispensable analytical discipline. It compels the writer to make value premises explicit rather than disguising them as factual conclusions, which sharpens accountability: a recommendation that "we ought to sanction" must state the operative principle, not merely the offending fact. In ethics case studies it provides a ready diagnostic for spotting where an argument illegitimately leaps from data to duty. Recognising Hume's Guillotine thus protects against rhetorical sleight of hand in committee reports, briefing notes, and editorial advocacy alike, ensuring that contested normative commitments are argued openly rather than smuggled past the reader.
Example
In its 2023 Sixth Assessment Report, the IPCC deliberately separated descriptive warming projections (the "is") from prescriptive mitigation targets (the "ought"), branding its advice "policy-relevant but not policy-prescriptive."
Frequently asked questions
The is-ought problem, from Hume's 1739 Treatise, is a logical claim that a prescriptive conclusion cannot be validly deduced from purely descriptive premises. G. E. Moore's naturalistic fallacy (1903) is a separate semantic claim that the property 'good' cannot be defined in terms of any natural property. One concerns inference; the other concerns conceptual analysis.
Keep learning